A relevance-based pairwise chromagram similarity for improving cover song retrieval accuracy

Jin Soo Seo

doi:10.7776/ASK.2024.43.2.200

All Issue

2024 Vol.43, Issue 2 Preview Page Next Page

Research Article

A relevance-based pairwise chromagram similarity for improving cover song retrieval accuracy 커버곡 검색 정확도 향상을 위한 적합도 기반 크로마그램 쌍별 유사도

31 March 2024. pp. 200-206

PDF XML

Abstract

Computing music similarity is an indispensable component in developing music search service. This paper proposes a relevance weight of each chromagram vector for cover song identification in computing a music similarity function in order to boost identification accuracy. We derive a music similarity function using the relevance weight based on the probabilistic relevance model, where higher relevance weights are assigned to less frequently-occurring discriminant chromagram vectors while lower weights to more frequently-occurring ones. Experimental results performed on two cover music datasets show that the proposed music similarity improves the cover song identification performance.

Keywords

Cover song identification

Music similarity

Probabilistic relevance model

Search weight

음악 유사도 계산은 음악 검색 서비스 구현에서 중요한 구성 요소 중 하나이다. 본 논문은 커버곡 검색 성능을 제고하기 위해서, 크로마그램 벡터 별로 커버곡 검색 적합도를 구하여 음악 유사도 계산 시 가중치로 활용한다. 커버곡 검색 적합도는 확률 적합도 모델을 이용하여 구한다. 커버곡 검색에 도움이 될 수 있는 분별도가 높은 벡터에 높은 가중치를 부여하고, 흔하게 존재하여 분별도가 떨어지는 벡터에 낮은 가중치를 부여하는 방식으로 음악 유사도 함수를 유도한다. 두 가지 커버곡 실험 데이터셋에서 성능 비교를 수행하여 제안한 음악 유사도 함수가 커버곡 검색 성능을 개선시킬 수 있음을 보였다.

키워드

커버곡 검색

음악 유사도

확률 적합도 모델

검색 가중치

References

F. Yesiler, G. Doras, R. M. Bittner, C. J. Tralie, and J. Serra, "Audio-based musical version identification: Elements and challenges," IEEE Signal Proc. Mag. 38, 115-136 (2021). 10.1109/MSP.2021.3105941

J. Serra, E. Gomez, P. Herrera, and X. Serra, "Chroma binary similarity and local alignment applied to cover song identification," IEEE Trans. Audio Speech Lang. Process, 16, 1138-1151 (2008). 10.1109/TASL.2008.924595

J. Seo, "A code-based chromagram similarity for cover song identification" (in Korean), J. Acoust. Soc. Kr. 38, 314-319 (2019).

F. Yesiler, J. Serra, and E. Gomez, "Accurate and scalable version identification using musically-motivated embeddings," Proc. ICASSP, 21-25 (2020). 10.1109/ICASSP40776.2020.9053793

T. F. Smith and M. S. Waterman, "Identification of common molecular subsequences," J. Mol. Biol. 147, 195-197 (1981). 10.1016/0022-2836(81)90087-57265238

J. Seo, "Pairwise similarity normalization based on a hubness score for improving cover song retrieval accuracy," IEICE Trans. Information and Systems, 105, 1130-1134 (2022). 10.1587/transinf.2021EDL8075

S. Robertson and K. S. Jones, "Relevance weighting of search terms," J. Am. Soc. Inf. Sci. 27, 129-146 (1976). 10.1002/asi.4630270302

S. Robertson, "Understanding inverse document frequency: on theoretical arguments for IDF," J. Documentation, 60, 503-520 (2004). 10.1108/00220410410560582

S. Robertson and S. Walker, "On relevance weights with little relevance information," Proc. SIGIR, 16-24 (1997). 10.1145/278459.258529

Covers80 Cover Song Data Set, Available, https://lab rosa.ee.columbia.edu/projects/coversongs/covers80/, (Last viewed March 12, 2024).

B. McFee and J. P. Bello, "Structured training for large-vocabulary chord recognition," Proc. ISMIR, 188-194 (2017).

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 43
No :2
Pages :200-206
Received Date : 2023-10-11
Accepted Date : 2024-01-09
DOI :https://doi.org/10.7776/ASK.2024.43.2.200

The Journal of the Acoustical Society of Korea ISSN:1225-4428(Print) 2287-3775(Online) 한국음향학회지

All Issue